www.nature.com/scientificreports

OPEN Dynamic variations in salinity and grade of a potassium‑rich deposit in Lop Nor basin, Lichun Ma1*, Kai Wang1,2, Yu Zhang1, Qingfeng Tang3 & Hui Yan4

The Quaternary Lop Nor playa is the largest production base of potassium sulfate in the world. It has a mining history of more than 10 years, and its share in the Chinese potassium sulfate market is about 50% to-date. In this basin, the high-salinity potassium-rich are mainly contained in Middle Pleistocene–Holocene glauberite strata. Based on the monitoring of the underground brine table and geochemical analysis, this study reveals variations in the underground brine table and potassium- bearing grade before and after large-scale mining in the Lop Nor deposit. The results showed that the underground brine table and potassium sulfate grade decreased by varying degrees over sub-mineral areas after large-scale mining. The underground brine table declined by 8.5 m, on average, in the Luobei depression, by 6.4 m in the Tenglong platform and by 1.9 m in the Xinqing platform. However, the potassium-bearing grade showed the diferent trend. The Tenglong platform

had the largest decline with average decreases in layers ­W1, ­W2 and ­W3 of 18.2%, 13.0% and 24.8%, respectively. In the Xinqing platform, the average decrease in ­layersW2 and ­W3 were 17.4% and 16.0% respectively. The Luobei depression decreases were relatively small ­(W1, ­W2 and ­W3 decreased 4.3%, 4.2% and 3.1%, respectively). This research provides a theoretical basis for the rational development and sustainable use of the potassium-rich brines in the Lop Nor basin.

Te Lop Nor playa is located in the eastern (, China) and is a famous Quaternary inland lake that is also the largest single liquid deposit of potassium sulfate in the world. Te high-salinity potassium- rich brine is mainly contained in Middle Pleistocene–Holocene glauberite strata. Te SDIC Xinjiang Lop Nor Potash Co., Ltd. (abbreviate: SLNP), founded in 2000, has the exploration and mining rights for the potash deposit. In 2003, SLNP carried out mining tests in the hinterland of Lop Nor basin. At the end of 2005, it reached an annual production capacity of 100,000 tons of potassium sulfate and in 2012, it achieved an annual output of 1.37 million tons. In 2017, it reached 1.5 million tons of production capacity. At present, its share in the Chinese potassium sulfate market is about 50%. Since the discovery of potassium-rich brine deposits in the Lop Nor basin, there has been a lot of research in the area. In the past 30 years, researchers have made great progress in metallogenic conditions and sedi- mentary environment analysis­ 1–6, ore deposit characteristic and genesis­ 2,7,8, geochemical characteristics of the potassium-rich brine reservoir body­ 2,9 and mining process and technology­ 10–13. However, there are few studies on the dynamic variations in the geochemistry of potassium-rich brine deposits in Lop Nor ­basin14. In 2006, 2009 and 2010, general exploration of all three sub-mineral areas (Luobei depression, and Tenglong and Xinqing platforms) in Lop Nor basin was carried out. Exploration obtained a large amount of drilling and brine chemistry data, which laid the foundation for this research. Based on the geochemical analysis of diferent ore horizons from the early general exploration ­reports15–17 and the brine samples and chemical data collected in the mining area in 2017, this study conducted a comprehensive comparison. It revealed the temporal and spatial variations of the brine mineralization and potassium grade in diferent ore horizons before and afer large-scale mining in Lop Nor basin. Our results provide a theoretical basis for the rational development and sustainable use of potassium-rich brines in Lop Nor basin.

1MNR Key Laboratory of Metallogeny and Mineral Assessment, Institute of Mineral Resources, Chinese Academy of Geological Science, Beijing 100037, China. 2China University of Geosciences(Beijing), Beijing 100083, China. 3Beijing Centre for Physical and Chemical Analysis, Beijing 100089, China. 4SDIC Xinjiang Lop Nor Potash Co., Ltd., 839000, China. *email: [email protected]

Scientifc Reports | (2021) 11:3351 | https://doi.org/10.1038/s41598-021-82958-y 1 Vol.:(0123456789) www.nature.com/scientificreports/

Figure 1. Simplifed geological map of the Lop Nor area in Xinjiang Province, China.

Regional geological background Lop Nor is located at the intersection of the Altun and Beishan tectonic belts in the eastern part of the Tarim platform. Te northern and southern basin boundary is controlled by the Kongque River fault and the Altun fault, ­respectively13 (Fig. 1). Te regional tectonic environment is complex, the basement is fractured and the inheritance and neotectonic activity are strong. Te fault structure generally restricts the formation and devel- opment of Lop Nor Salt ­Lake2. Since the end of the Neogene, the Lop Nor area has been controlled by tectonic movement­ 5; it began to settle and gradually developed into the lowest depression of the basin. Glacial meltwater, originating from the Tianshan, Kunlun and Altun Mountains, eventually gathered in the Lop Nor basin­ 13; the water area once reached 20,000 km2. Terefore, the Lop Nor Lake played an important role as a catchment center of the whole Tarim basin throughout the Quaternary by accumulating a large amount of salt in the ­basin13. At present, the lake has completely dried up with a salt cover spanning 10,000 km2; the salt deposit is more than 200 m thick. Potassium-rich brines are found in these Quaternary salt strata in the northern Lop Nor basin. Te potassium-rich brine deposit consists of three sub-mineral areas from west to east, including the Xin- qing platform, Luobei depression and Tenglong platform. Te boundary between the sub-mines and the brine reservoir is mainly controlled by faults. According to the fault direction, the system can be roughly divided into three groups: faults in the NNE, faults near the EW and faults in the NEE. Te faults in the NEE are the most developed and include the F4 fault, which is between the Xinqing mining area and the Luobei depression, and the F6 fault, which is the boundary of the Luobei depression and the Tenglong platform­ 14. Te F1 fault is a regional compression–torsion fault, which passes through the Tenglong mining area and cuts it into two parts, north and ­south14 (Fig. 1). Luobei depression is located in the middle of the whole mining area. It is about 60 km long from north to south, 32.5 km wide from east to west, and has an area of about 1534 km2 (Fig. 1). Te overall terrain is lower than the Tenglong and Xinqing platforms to the east and west with an average altitude of about 780 m, and the surface is covered by salt crusts. Te thickness of the salt strata ranges from about 30–200 m, and the average thickness is about 100 m. Te salt system tends to thicken from south to north and the thickness also gradually increases from west to ­east15. Te Tenglong platform is about 70–90 km long from north to south and about 20–25 km wide from east to west with an area of about 1623 km2 (Fig. 1). Te surface is mainly a Yadan landform and the terrain is higher than the Luobei depression, with a maximum altitude of 790 m and a minimum altitude of 780 m (aver- age = 785 m). Te thickness of the salt-bearing strata in the northern part of the Tenglong platform is generally 30–50 m (average = 36 m) with a maximum thickness of 69.6 m. Te thickness gradually increases from north

Scientifc Reports | (2021) 11:3351 | https://doi.org/10.1038/s41598-021-82958-y 2 Vol:.(1234567890) www.nature.com/scientificreports/

(12 m) to south (60 m) and from east (20 m) to west (60 m). Te thickness of the salt-bearing strata is generally 5–20 m south of the F1 fault, and gradually increases from south to north (4–12 m) and decreases from east to west with a thick middle ­area16. Xinqing platform is about 60–80 km long (north–south) and 10–20 km wide (east–west) with an area of about 1447 km2 (Fig. 1). Te surface is mainly a Yadan landform and the terrain is relatively high (highest altitude = 795 m, lowest altitude = 780 m) with an average elevation of 789 m. Te accumulated thickness of the salt-bearing strata in the Xinqing platform is about 30–50 m. It gradually thickens from south to north, as well as west to east­ 17. Te whole Lop Nor playa currently has no surface water system. It mainly receives bedrock fssure water and groundwater recharge from Kuluktag Mountain, Beishan Mountain and the Altun Mountains. It also receives lateral recharge from the eastern Archik Valley, as well as the western Kongque River and dry deltas (Fig. 1). Down the regional hydrologic gradient, as groundwater moves from the outer fringes to the center of the playa, its salinity gradually increase, reaching a maximum of roughly 350 g/L. Methods Sampling and analysis. In 2003, SLNP conducted a mining test in the Lop Nor playa. Te underground brine was pumped to the surface through a shaf system, and then transported through a brine channel to the solar pond for salt drying and classifcation. Following an economic feasibility analysis, the current four-layer brine (top to bottom: ­W1, ­W2, ­W3 and W­ 4) was mainly mined within 90 m. However, the amount mined early on was relatively small. At the end of 2005, the production capacity was only 100,000 tons ­(K2SO4) per year. Te potassium sulfate project goal of 1.2 million tons per year started ran until November 2008. In November 2011, the monthly design production capacity was reached. Tis meant that large-scale brine mining had not been carried out in the Lop Nor playa before 2011. To fnd the potash reserves in the mining area, SLNP carried out general exploration operations in July 2006, March 2009 and August 2010 in the Luobei depression, Tenglong platform and Xinqing platform, respectively. Te exploration data provided the basis for the study of geochemi- cal variations before and afer large-scale mining in the Lop Nor basin. Te chemistry analysis from the three sub-mines areas used in this study was derived from the general exploration reports from 2006, 2009 and ­201015–17. Data were generated from 88 brine samples collected in July–August 2017 through observation holes in the mine area. Two bottles (500 mL each) were used to sample each observation hole. Te brine depth and density were measured on site and the bottles were sealed quickly afer measurements to prevent the brine from evaporating or leaking during transport. GPS data were used for geolocation and elevation measurements. Te brine samples taken in 2017 were sent to the National Geological Experiment and Test Center (Chi- + + 2+ 2+ − 2− nese Academy of Geological Sciences) for the analysis of major components (Na­ , ­K , ­Mg , ­Ca , ­Cl , ­SO4 , 2− − + 3+ − − + + 2+ + + 2+ 2+ + 3+ ­CO3 and ­HCO3 ) and trace elements ­(Li , ­B , ­Br , ­I , ­Rb , ­Cs and ­Sr ) in which ­K , Na­ , ­Mg , ­Ca , ­Li , ­B , + + 2+ − 2− ­Rb , ­Cs and Sr­ were determined by atomic absorption spectrophotometry (RSD < 2%), Cl­ and SO­ 4 were determined by ion-chromatography (RSD < 2%). Titrimetric methods were used for the determination of Br­ −, − 2− − ­I , ­CO3 and ­HCO3 (RSD < 5%).

Mapping methods. According to the brine chemistry analysis and the groundwater table data, a spatial distribution map of brine geochemistry and groundwater drawdown in diferent ore horizons of the three sub- mining areas were drawn by Kriging interpolation in the sofware Surfer. Results and discussion Lop Nor is a typical liquid deposit of potassium sulfate. Based on the Valyashko classifcation system­ 18, the water chemistry type is mainly a sulfate subtype and, secondly, a sodium sulfate subtype. Te current brine salinity range is about 226–393 g/L and the average grade of KCl is about 1.36%. Te brine reservoir is mainly located in the glauberite layer, then the coarse clastic layer and a very small amount is in the halite and gypsum layer. Te number of potassium-rich brines layers is diferent in each sub-mineral area, which is controlled by the structure and fault system of the mining area. Tere are seven brine layers (W­ 1–W7) within 250 m depth in the Luobei depression, including one layer of phreatic water and six layers of confned water (Fig. 2). Te Xinqing mining area has two layers of confned water (Fig. 2). Te Tenglong mining area has three potassium-rich brine layers, including a phreatic water layer and two confned water layers (Fig. 2). Te Location of cross-section A–A′ is shown in Fig. 1. According to analyses of principal and trace elements from the brine samples collected in 2017, the available elements in the brine of the Lop Nor playa included elemental B in addition to K, and the content of ­B2O3 in the brine varied from 277.3 to 755.6 mg/L, which is greater than the comprehensive utilization grade (150 mg/L). However, B has not been exploited yet, so it will not be discussed in this study.

Luobei depression. Luobei depression is the main storage area for potassium-rich brine. Trough explora- tion in 2006, it was found that the KCl (122b + 333) specifc yield reserve in the elevation range of 628–786 m was 8384.84 million ­tons15, and long-term observation holes were established for monitoring variations in chemical and physical properties and the water table of the underground brine in three main ore layers (W1, W2, W3). In 2006, the burial depth of the ore layer ­W1, ­W2, ­W3 and ­W4 was 1.7–2.3 m, 20–40 m, 40–60 m and 55–75 m respectively. Afer 11 years of mining, the brine table of layer W­ 1 decreased by 8.5 m on average. Figure 3 shows the comparison of the salinity in three ore layers from 2006 and 2017, including the temporal and spatial varia- tions in brine salinity before and afer 11 years of mining in the Luobei depression.

Scientifc Reports | (2021) 11:3351 | https://doi.org/10.1038/s41598-021-82958-y 3 Vol.:(0123456789) www.nature.com/scientificreports/

Figure 2. Cross-section of the potassium-rich brine deposit in Lop Nor basin.

In 2006, the salinity of the ore layer W­ 1 in the Luobei depression was 295.9–387.8 g/L and the average salinity was 348.8 g/L15 (Fig. 3a). Te low-salinity area is located in the northern part of the mining area with higher values located in the south. Afer 11 years of mining, the salinity ranged from 234.7 to 387.4 g/L and was signifcantly reduced in the northern and southwestern regions, which may have been caused by the recharge of freshwater from the northern piedmont zone and the western dry delta. Te high-salinity area is ofset to the northeast, which means that the underground brine formed a new fow feld and the concentration center moved to the Tenglong platform (Fig. 3b). Ore layers W­ 2 and W­ 1 showed the same trends; in 2006, the salinity in ­W2 was 310.3–387.6 g/L, but in 2017, the salinity was changed to 316.9–367.7 g/L. Te concentration center also migrated from the south–central part of the mining area to the east. Afer years of mining, the brine in the north, west and southwest is obviously depleted (Fig. 3c, d). Te salinity of layer W­ 3 in the south was also higher than in the north. Te highest value was 389.4 g/L and the lowest value was 307.2 g/L with an average salinity of 352.1 g/L (Fig. 3e). Afer mining, the salinity varied between 344.5–380.6 g/L and the high-salinity area moved to the north where brine salinity increased obviously. Tis may be because of high-salinity brine recharge from the eastern and western platforms afer the formation of new fow felds. As the depth increased, W­ 3 was not signifcantly afected by freshwater from the northern piedmont zone and western deltas (Fig. 3f). Te spatial distribution of ­K+ content in three ore layers in 2006 and 2017 is shown in Fig. 4. In 2006, the K­ + content of ­W1 in the Luobei depression increased from 4.41 to 13.4 g/L with an average content of 9.4 g/L. Te main high-value area is located in the southern part of the mining area. Te ­K+ content in the northern part of the mining area was relatively low (5.5–8 g/L), which may be related to the recharge of the bedrock fssure water from Kuluktag Mountain in the north(Fig. 4a). Afer 11 years of mining, in 2017, the ­K+ content changed to 5.9–10.9 g/L (average = 9.0 g/L), which was an average decrease of 4.3%. Te main high-value area shrunk and was located in the central and northeastern parts of the mining area (Fig. 4b). + + In 2006, the K­ content of ­W2 was 6.9–12.3 g/L with an average of 9.5 g/L. Te K­ distribution was similar to ­W1. Te high-value area is mainly located in the southern part of the mining area, and the low-value area is located in the north (Fig. 4c). Afer mining, the K­ + content was 6.67–10.3 g/L (average = 9.1 g/L), which was a decline of about 4.2%, and the high-value area shrunk from the south–central part of the mining area to the + southeastern part (Fig. 4d). Before mining ore layer ­W3, the distribution of ­K was similar to ­W1 and ­W2 in that it was also characterized by a low-value north and a high-value south. Te ­K+ content varied from 6.9 to 12.4 g/L with an average of 9.8 g/L (Fig. 4e). Afer mining, the K­ + content generally changed to 8.2–10.4 g/L and the aver- age content decreased from 9.8 to 9.5 g/L, a decline of 3.1% (Fig. 4f).

Tenglong platform. Te Tenglong mining area exposed three layers of potassium-rich brine, including a phreatic water layer and two confned water layers. Te phreatic brine ore body is the main ore body of the Ten- glong mining area. It is buried at shallow depth with the water table at about 0.7–4.0 m. However, the phreatic brine ore body is bounded by a fracture (F1) and only present in the northern part of the Tenglong mining area. Te burial depth of the confned brine layer ­W2, ­W3 was about 15–30 m, 20-40 m respectively. In March 2009, exploration determined a specifc yield reserve of KCl (122b + 333) of 26.53 million tons at 707–787 m elevation­ 16. Figure 5 shows a comparison of the salinity of three ore layers in 2009 and 2017, specif- cally the temporal and spatial variations in salinity before and afer large-scale mining in the Tenglong mining area. Te salinity of horizon ­W1 in the Tenglong mining area in 2009 was 344.5–390.1 g/L with an average value of 365.3 g/L16 (Fig. 5a). Te high-salinity area is mainly located in the middle of the mining area and the low-value area is located in the southwest and north. Afer 8 years of mining, the salinity trends were basically the same, the values changed from 329.2 to 379.4 g/L (Fig. 5b). In 2009, the salinity of horizon W­ 2 changed from 342.7 to 391.5 g/L (average = 366.9 g/L). Te high-value area was distributed in the east–central and south areas, and low salinity is mainly concentrated in the northern part of the mining area (Fig. 5c). In 2017, afer mining, the salinity range changed to 331.2–387.7 g/L and the southeastern part of the mining area was obviously diluted

Scientifc Reports | (2021) 11:3351 | https://doi.org/10.1038/s41598-021-82958-y 4 Vol:.(1234567890) www.nature.com/scientificreports/

Figure 3. Temporal and spatial variations in the salinity of diferent brine layers in Luobei depression. (a) Layer ­W1 in 2006, (b) Layer W­ 1 in 2017, (c) Layer W­ 2 in 2006, (d) Layer W­ 2 in 2017, (e) Layer W­ 3 in 2006 and (f) Layer W­ 3 in 2017. Te green dots indicate observation holes locations.

Scientifc Reports | (2021) 11:3351 | https://doi.org/10.1038/s41598-021-82958-y 5 Vol.:(0123456789) www.nature.com/scientificreports/

Figure 4. Temporal and spatial variations in ­K+ content of diferent brine layers in Luobei depression. (a) Layer ­W1 in 2006, (b) Layer W­ 1 in 2017, (c) Layer W­ 2 in 2006, (d) Layer W­ 2 in 2017, (e) Layer W­ 3 in 2006 and (f) Layer W­ 3 in 2017. Te green dots indicate observation holes locations.

Scientifc Reports | (2021) 11:3351 | https://doi.org/10.1038/s41598-021-82958-y 6 Vol:.(1234567890) www.nature.com/scientificreports/

Figure 5. Temporal and spatial variations in salinity of diferent brine layers in Tenglong platform. (a) Layer W­ 1 in 2009, (b) Layer W­ 1 in 2017, (c) Layer W­ 2 in 2009, (d) Layer W­ 2 in 2017, (e) Layer W­ 3 in 2009 and (f) Layer ­W3 in 2017. Te black crosses indicate observation holes locations.

Scientifc Reports | (2021) 11:3351 | https://doi.org/10.1038/s41598-021-82958-y 7 Vol.:(0123456789) www.nature.com/scientificreports/

(Fig. 5d). Horizons W­ 3, ­W1 and W­ 2 showed a consistent distribution trend before and afer mining. Te post- mining salinity range was reduced from 343.9–397.7 to 290.9–353.7 g/L, a decline of 8.4% (Fig. 5e, f). Figure 6 compares the spatial distribution of K­ + content before and afer mining in the Tenglong platform. + Te K­ content in ­W1 in 2009 changed from 8.9 to 12.9 g/L with an average content of 11.0 g/L (Fig. 6a). Afer mining, the whole area showed a downward trend. Te K­ + content changed to 8.2–9.8 g/L with an average content + of 9.0 g/L, a drop of 18.2% (Fig. 6b). In 2009, the K­ content of horizon W­ 2 was 4.3–13.1 g/L (average = 10.8 g/L). Te high-value area was distributed in the south–central and northern parts of the mining area (Fig. 6c). Afer mining, the ­K+ content changed to 8.3–10.3 g/L (average = 9.4 g/L), an average reduction of 13%, especially in the + northern and southeastern parts of the mining area (Fig. 6d). In 2009, the spatial distribution of K­ in W­ 3 was + + similar to ­W2. Te ­K content varied from 7.7 to 12.3 g/L with an average of 10.9 g/L (Fig. 6e). Te ­K content afer mining ranged from 6.5 to 9.3 g/L (average = 8.2 g/L), which represented a drop of 24.8%; the largest decline was in the north and southeast (Fig. 6f).

Xinqing platform. Te study revealed two layers of potassium-rich brine ore, both of which are confned water layers. General exploration was carried out in August 2010, which reported a KCl (122b + 332 + 333) spe- cifc feld reserve of 7.60 million tons in elevations ranging from 742 to 800 m17. Te burial depth of the ore layer ­W2, ­W3 was about 10–20 m, 11.3–38 m respectively. Salinity comparisons in 2017 are shown in Fig. 7. 17 In 2010, the salinity of ­W2 in Xinqing platform was 268.6–374.0 g/L and the average value was 345.9 g/L (Fig. 7a). Te high-salinity area is mainly located in the middle and eastern area, while the low-salinity area is located in the west and results from the lateral recharge of the Tarim River delta. Afer 7 years of mining, the whole mining area was desalinated and the salinity changed to 258.7–351.0 g/L, particularly in the southeastern part (Fig. 7b). Te changes in the W­ 3 layer before and afer mining are consistent with W­ 2. Te mineraliza- tion degree of ­W3 in 2010 was 251.6–381.7 g/L (average = 333.5 g/L). Afer mining, the salinity decreased to 226.1–366.1 g/L. Te high-value area moved to the northeast and the southeastern area faded signifcantly (Fig. 7c, d). At the same time, the ­K+ content also changed accordingly. Te spatial distribution of ­K+ in 2010 and 2017 is shown in Fig. 8. + Te ­K content of W­ 2 in 2010 was 4.8–12.4 g/L and the average content was 9.8 g/L (Fig. 8a). Te area of high K­ + content is mainly located in the north–central region of Xinqing platform. In the northern mining area, there were very few samples with the low values, which may have been caused by the replenishment of bedrock fssure water of the Kuluktag. Afer 7 years of mining, the ­K+ content decreased signifcantly; the ­K+ content was 6.3–10.0 g/L (average = 8.1 g/L), which represents an average drop of 17.4%. Te northeastern area was the only region to exhibit high values, while most of the western, central and southern regions were signifcantly diluted (Fig. 8b). + Figure 8c shows the spatial distribution characteristics of the K­ content in W­ 3 in 2010. Similar to the dis- + + tribution of ­W2, the ­K content varied from 4.5 to 12.5 g/L (average = 9.4 g/L). Afer 7 years of mining, the K­ content decreased by16% to 4.4–9.0 g/L (average = 7.9 g/L). Te K­ + content of the whole mining area decreased signifcantly, especially in the southern part, with declines as high as 50% (Fig. 8d).

Underground brine drawdown. To describe variations in the underground brine table before and afer large-scale mining of potassium-rich brine deposits in Lop Nor basin, this study used early monitoring data from three sub-mine areas. Te data showed that the thickness of reservoir ­W1 in the Luobei depression was about 15–25 m in 2006. Te burial depth was generally 1.7–2.3 m with a maximum of 4.5 m15. Afer mining, the brine table dropped signifcantly. From 2006 to 2017, the maximum brine drawdown reached 16.5 m (mini- mum = 0.2 m, average = 8.5 m; Fig. 9a). Te Tenglong mining area also experienced a large drop in the water table. In 2009, the average thickness of reservoir W­ 1 in the northern part of the Tenglong mining area was 30–50 m, and the burial depth was about 0.7–4.0 m­ 16. Te 2017 monitoring data showed that the maximum drawdown was 13.6 m and the minimum drawdown was 0.2 m (average = 6.4 m; Fig. 9b). Tere is no phreatic brine in the Xinqing platform. Te shallowest confned brine layer is ­W2. In 2010, the ­W2 ore body was buried at a depth of 10–20 m with a thickness of 2–5 m17. Afer 7 years of mining, the reduction was small compared with a phreatic brine layer; the maximum drawdown was 4.3 m, the minimum was 0 and the average was 1.9 m (Fig. 9c). In general, the Luobei depression had the largest decline of the underground brine table and the drawdown of the water level in the most areas was greater than 10 m. Additionally, a settlement funnel was formed, centered on Luobei depression (Fig. 10). In summary, the factors controlling the brine table drawdown include recharge (lateral subsurface recharge around the basin, atmospheric precipitation, condensate, and possible recharge from deep source) and discharge (artifcal exploitation and evaporation). When the discharge is greater than the recharge, the groundwater table will continue to drop. However, the variations of salinity and potassium content are more complicated. It is not only controlled by the balance of recharge and discharge, but also involves the salt dissolution from evaporate strata and solute circulation in new fow feld. Nevertheless, the artifcial exploitation is still the main factor to control the variations in salinity and potassium content. It is estimated that the amount of underground brine extracted is ~ 29,600 × 104 m3/a, and the total amount of recharge is only ~ 6700 × 104 m3/a in the Lop Nor playa­ 15. Te diference between extraction and recharge is about 22,900 × 104 m3/a. If the brine salinity is about 300 g/L (0.3 ton/ m­ 3), it is estimated that more than 6500 × 104 tons of salt were exported every year. Terefore, the brine table, salinity and potassium content have decreased signifcantly afer more than 10 years of exploitation.

Scientifc Reports | (2021) 11:3351 | https://doi.org/10.1038/s41598-021-82958-y 8 Vol:.(1234567890) www.nature.com/scientificreports/

Figure 6. Temporal and spatial variations in K + content of diferent brine layers in Tenglong platform. (a) Layer W­ 1 in 2009, (b) Layer W­ 1 in 2017, (c) Layer W­ 2 in 2009, (d) Layer W­ 2 in 2017, (e) Layer W­ 3 in 2009 and (f) Layer W­ 3 in 2017. Te black crosses indicate observation holes locations.

Scientifc Reports | (2021) 11:3351 | https://doi.org/10.1038/s41598-021-82958-y 9 Vol.:(0123456789) www.nature.com/scientificreports/

Figure 7. Temporal and spatial variations in salinity of diferent brine layers in Xinqing platform. (a) Layer ­W2 in 2010, (b) Layer W­ 2 in 2017, (c) Layer W­ 3 in 2010 and (d) Layer W­ 3 in 2017. Te red crosses indicate observation holes locations.

Conclusions China is a large agricultural country with a potash defciency. At present, the self-sufciency rate of potash fertilizer is only 50%. Te Quaternary Lop Nor playa is one of the most important potash deposits in China. Although its mining history is not long, it supplies 50% of the Chinese potassium sulfate market. Afer more than 10 years of mining, the underground brine table and the grade of potassium sulfate in each sub-mine of the Lop Nor potash deposit have declined to varying degrees. Afer large-scale mining, the ­K+ content in the Tenglong mining area decreased the most, followed by Xinqing + and then Luobei. Te average K­ content of horizon W­ 1 was 9.0 g/L in the Tenglong area, which represented an + + average decrease of 18.2%. Te average K­ in W­ 2 was 9.4 g/L, indicating a decrease of 13%, and the average K­ + content in ­W3 was 8.2 g/L (24.8% decline). Afer mining in the Xinqing mining area, the average K­ content in + layer ­W2 was 8.1 g/L, with an average decline of 17.4%. Te average ­K content in layer ­W3 was 7.9 g/L with a

Scientifc Reports | (2021) 11:3351 | https://doi.org/10.1038/s41598-021-82958-y 10 Vol:.(1234567890) www.nature.com/scientificreports/

Figure 8. Temporal and spatial variations in K + content of diferent brine layers in Xinqing platform. (a) Layer ­W2 in 2010, (b) Layer W­ 2 in 2017, (c) Layer W­ 3 in 2010 and (d) Layer W­ 3 in 2017. Te red crosses indicate observation holes locations.

decline of 16.0%; in the southern part of the mining area, the decline was as high as 50%. Afer mining in Luobei + + depression, the average K­ content of ­W1 was 9.0 g/L, showing an average reduction of 4.3%. Te average K­ content in ­W2 was 9.1 g/L (4.2% decline) and 9.5 g/L (3.1% decline) in W­ 3. Te long-term monitoring data of the underground brine table from observation holes showed that the brine table of the Luobei depression obviously decreased afer mining, with a maximum drawdown of ­W1 brine reach- ing 16.5 m (average = 8.5 m). A large drop in the brine level also occurred in the Tenglong mining area where the maximum drawdown of W­ 1 was 13.6 m (average = 6.4 m). Xinqing platform did not contain phreatic brine. Terefore, compared with a phreatic brine layer, the resulting reduction was small and the lowest, shallowest confned layer ­(W2) had a relatively small reduction (maximum drawdown = 4.3 m, average = 1.9 m). Te artifcial exploitation is the main factor to control the brine table drawdown, the variations in salinity and potassium content. According to the existing exploitation scale (~ 23,000 × 104 m3/a), the service life of phreatic brine layer W1 is about 10 years. Te resource reserves of confned brine body have not been identifed, and further exploration is quite necessary to ensure the long-term sustainable utilization of the potassium-rich brines resources in the Lop Nor basin.

Scientifc Reports | (2021) 11:3351 | https://doi.org/10.1038/s41598-021-82958-y 11 Vol.:(0123456789) www.nature.com/scientificreports/

Figure 9. Te drawdown of potassium-rich brine deposits in Lop Nor basin. (a) Te depth of brine drawdown in Luobei depression. (b) Te depth of brine drawdown in Tenglong mining area, (c) Te depth of brine drawdown in Xinqing platform. Te abscissa is the observation well number.

Figure 10. Te drawdown distribution of potassium-rich brine layer W­ 1 in Luobei depression and Tenglong mining area. Te unit of brine drawdown is meter. Te red line indicates the fault.

Received: 9 December 2019; Accepted: 19 January 2021

References 1. Zheng, M. P., Qi, W., Wu, S. Y. & Liu, J. Y. A preliminary study on the sedimentary environment and prospect of potassium explora- tion in Lop Nur since the late pleistocene. Chin. Sci. Bull. 36(23), 1810–1810 (1991). 2. Wang, M. L. Potash Resources in Lop Nor Salt Lake 1–386 (Geological Publishing House, Beijing, 2001). 3. Zhao, Z. H. et al. Discussion of the lower limit of quaternary in Lop Nur, Xinjiang. J. Arid Land Geogr. 2, 130–135 (2001).

Scientifc Reports | (2021) 11:3351 | https://doi.org/10.1038/s41598-021-82958-y 12 Vol:.(1234567890) www.nature.com/scientificreports/

4. Hu, D. S. & Zhang, H. J. Lake-evaporated salt resources and the environmental evolution in the Lop Nur Region. J. Glaciol. Geocryol. 26(2), 212–218 (2004). 5. Lin, J. X. et al. Te lithostratigraphy, magnetostratigraphy, and climatostratigraphy in the Lop Nur region. J. Stratigr. 29(4), 317–322 (2005). 6. Ma, L. C., Liu, C. L., Jiao, P. C. & Chen, Y. Z. A preliminary discussion on geological conditions and indicator pattern of Potash deposits in typical Playas of Xinjiang. Miner. Depos. 29(4), 593–601 (2010). 7. Liu, C. L. et al. Characteristics of diagenesis of the quaternary salt-bearing strata, Lop Nur Lake, Xinjiang. Acta Geosci. Sin. 21(2), 240–246 (2003). 8. Liu, C. L., Wang, M. L. & Jiao, P. C. Te probing of regularity and controlling factors of potash deposits distribution in Lop Nur Salt Lake, Xinjiang. Acta Geosci. Sin. 30(6), 796–802 (2009). 9. Liu, C. L., Wang, M. L., Jiao, P. C., Cheng, Y. Z. & Li, S. D. Formation of pores and brine reserving mechanism of the aquifers in quaternary potash deposits in Lop Nur Lake, Xinjiang, China. Geol. Rev. 48(4), 437–443 (2002). 10. Cheng, Y. Z., Wang, M. L., Yang, Z. S., Liu, C. L. & Jiao, P. C. Te making of potash-bearing mixtures through the processing of magnesium sulfate sub-type Brine in Lop Nur Saline Lake, Xinjiang. Acta Geosci. Sin. 22(5), 465–470 (2001). 11. Gu, X. L., Lu, C. X., Zeng, Y. G. & Song, W. J. Quaternary salt mineral of containing potassium occurrence character analyzing in Lop Nor Tonlon Terrace. Xinjiang Geol. 27(04), 346–349 (2009). 12. Gu, X. L. et al. Analysis of the developing prospect of the unconfned brine kalium mine in the north hollow of the Lop Nur Region. Hydrogeol. Eng. Geol. 30(02), 32–36 (2003). 13. Ma, L. C., Lowenstein, T. K., Li, B. G., Jiang, P. A. & Wu, H. Q. Hydrochemical characteristics and brine evolution paths of Lop Nor Basin, Xinjiang Province, Western China. Appl. Geochem. 25(11), 1770–1782 (2010). 14. Sun, M. G. & Ma, L. C. Potassium-rich brine deposit in Lop Nor basin, Xinjiang, China. Sci. Rep. 8(7676), 1–9 (2018). 15. SDIC Xinjiang Lop Nor Potash Co., Ltd. General Exploration Report on the Potash Deposit in Luobei Depression, , Xinjiang 1–295 (SLNP Press, Hami, 2006). 16. SDIC Xinjiang Lop Nor Potash Co., Ltd. General Exploration Report on the Potash Deposit in Tenglong Platform, Ruoqiang County, Xinjiang 1–266 (SLNP Press, Hami, 2009). 17. SDIC Xinjiang Lop Nor Potash Co., Ltd. General Exploration Report on the Potash Deposit in Xinqing Platform, Ruoqiang County, Xinjiang 1–251 (SLNP Press, Hami, 2011). 18. Valyashko, M. G. Geochemical Regularity of Potash Deposit Formation 1–353 (China Industry Press, Beijing, 1965). Acknowledgements This work was funded by subject of National Key Research and Development Program of China (No. 2017YFC0602806), Central Public Interest Scientifc Institution Basal Research Fund (No. JYYWF2018) and China Geological Survey project (No. DD20160054). We are grateful to Mr. Wenxue Li, Mr. Baocheng Ma and Mr. Baoheng Yang from SDIC Xinjiang Lop Nor Potash Co., Ltd for their help with feld work. We thank Kara Bogus, Ph.D., from Liwen Bianji, Edanz Editing China (www.liwen​bianj​i.cn/ac), for editing the English text of a draf of this manuscript. Author contributions L.M. designed the research in the manuscript. L.M., K.W., Y.Z., Q.T. and H.Y. participated in sample collection and processing. L.M. and K. W. wrote the frst draf and prepared all fgures. All authors reviewed the manuscript.

Competing interests Te authors declare no competing interests. Additional information Correspondence and requests for materials should be addressed to L.M. Reprints and permissions information is available at www.nature.com/reprints. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional afliations. Open Access Tis article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. Te images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creat​iveco​mmons​.org/licen​ses/by/4.0/.

© Te Author(s) 2021

Scientifc Reports | (2021) 11:3351 | https://doi.org/10.1038/s41598-021-82958-y 13 Vol.:(0123456789)